Introduction
Since the epic explorations of the early twentieth century, there has been a steadily growing human presence in Antarctica. Initially almost all visits were made in summer but, more recently, increasing numbers of people have overwintered. The first permanent base was established on an island in 1904 (W.S. Bruce's station on Laurie Island, later taken over by Argentina) while, on the Antarctic continent, the base with the longest record of continuous occupation (since 1954) is Australia's Mawson station (Bleasel and others Reference Bleasel, Bonner, Bolin and Knox1989; Headland Reference Headland1989: 229). By 1988 there were 43 continuously inhabited stations, 32 permanent summer bases and significant numbers of field camps, with a total complement in summer of about 3000 personnel (Bleasel and others Reference Bleasel, Bonner, Bolin and Knox1989). Since then, the establishment of new stations appears virtually to have ceased, with Hughes (Reference Hughes2004) reporting only 44 continuously occupied bases. Nevertheless, the number of people visiting the Antarctic is steadily growing, and at present around 5000 people from at least 29 nations spend all or part of the year working there (Heaton and Paterson Reference Heaton and Paterson2003).
Wherever people go, human wastes are produced and have to be disposed of. Early visitors to Antarctica did what they had to do, as quickly as possible, and moved on. However, once people began to spend long periods in one place, aesthetic and hygienic considerations made a more organised approach to human waste disposal essential. Although a number of countries have made territorial claims in Antarctica, these have no formal international recognition and so no country or body has the legal power to prescribe how human wastes should be dealt with. Nevertheless, those countries that are signatories to the Antarctic Treaty have collaborated to produce guidelines on how wastes of various kinds, including sewage, should be disposed of. The most recent of these, the so-called Madrid Protocol (see below) has prompted a number of countries to forsake well-established (and still acceptable) methods of disposal and install modern wastewater treatment plants.
The first aim of this paper is to examine how wastewater treatment practices in Antarctica have evolved, and what the drivers for change have been. Its second aim is to review the types of treatment plants that have been installed at research stations and how these choices have been determined by the many constraints affecting engineering projects in the Antarctic.
Drivers for change
As mentioned above, considerations of hygiene, aesthetics and convenience largely determined how waste was disposed of in the early days in Antarctica. For example, Kerry (Reference Kerry2002) describes how collected human urine at Australia's coastal stations was tipped into the sea while sludge was sea-iced (taken out onto the sea ice during winter and dispersed into the ocean when the sea ice broke up and drifted away in summer). From the 1960s onwards, there has been a growing worldwide awareness of, and concern about, the environment and this led to the introduction, at the eighth Antarctic Treaty Consultative Meeting (ATCM VIII) in 1975, of a Code of Conduct for Antarctic Expeditions and Station Activities (Bleasel and others Reference Bleasel, Bonner, Bolin and Knox1989). This stated that coastal stations should macerate their sewage before discharging it to the sea, preferably into deep waters. For inland stations, return of wastes to supporting stations or ships was recommended; however if this were impracticable, disposal into deep ice pits was the preferred disposal option. These guidelines were effectively reaffirmed in guidelines on solid waste disposal introduced at the eighteenth meeting of the Scientific Committee on Antarctic Research (SCAR XVIII) in 1984 (Bleasel and others Reference Bleasel, Bonner, Bolin and Knox1989).
In the late 1980s, there was a concerted push to provide greater protection for the Antarctic environment. Out of this came the Protocol on Environmental Protection to the Antarctic Treaty, the so-called Madrid Protocol, which was adopted by the Antarctic Treaty nations in 1991. Annex III to this protocol deals with waste disposal and waste management. Article 1 of this annex states that, as a general principle, wastes should be removed from Antarctica by those generating them. Article 4 recommends that, as far as is practicable, sewage from coastal stations should not be disposed of to sea ice, ice shelves or grounded ice sheets. Inland stations are permitted to dispose of sewage in deep ice pits if local disposal is the only practicable option. However, such disposal should not be undertaken in iceflow lines terminating in ice free areas or areas of high ablation. Article 5 reaffirmed that sewage and the products of wastewater treatment plants could be disposed of at sea but recommended that bases housing more than 30 people should subject their wastes to further treatment (with maceration as the minimum requirement) before discharge (Australian Antarctic Division 2005).
Responses to this protocol have varied. It would appear that a few countries have yet to comply with all its provisions; for example a recent report of the Council of Managers of National Antarctic Programmes (CONMAP) mentions instances in which sewage was still being discharged to ice-free land, in contravention of the protocol's stipulations (CONMAP 2002). Many countries have remained content to comply with the minimum requirements for sewage disposal. At France's Dumont d'Urville station, for example, the authorities have opted to continue discharging untreated sewage to the sea, as has been done for the past 40 years (Delille and Delille Reference Delille and Delille2000). However, for a number of countries, the Madrid Protocol has provided a strong stimulus to introduce more advanced sewage treatment methods.
Another strong driver for change has been the growing concern about the impact of sewage discharges on the Antarctic environment, both in the short- and long-terms. A 1980 review of environmental impacts at the American McMurdo and Palmer stations raised questions about the impact of heavy metals and suspended solids in sewage discharges on benthic organisms. The issue of toxic compounds was also raised (Division of Polar Programs 1984).
The 1989 SCAR report noted the limited level of microbial activity at the low temperatures prevalent in Antarctica. Waste decomposition rates are low and there is potential for waste to persist for a long time in an undecomposed state. Particular concern was expressed about the discharge of waste in ice-free areas, which are very scarce in Antarctica and which support soil communities of great scientific interest. Once contaminated, such communities are permanently affected (Bleasel and others Reference Bleasel, Bonner, Bolin and Knox1989). The effect of the low Antarctic seawater temperatures has been quantified by Howington and others (Reference Howington, McFeters, Jones and Smith1994) who showed that at –1.8°C the extent of BOD (biochemical oxygen demand; a measure of a wastewater's content of biodegradable organic compounds) removal was only one third of that at 20°C. The fate of pathogens under Antarctic conditions has also been explored and Parker and Martel (Reference Parker and Martel2002) concluded that some bacteria, viruses and helminth ova could survive for long periods in a frozen state.
Concerns about the impacts of coastal sewage discharges have led to the mapping of sewage plumes around outfalls from a number of research stations. Around Dumont d'Urville station relatively high densities of bacteria (up to 1000 cfu/100 ml) were found in the immediate vicinity of the outfall but bacterial indicators dropped to very low levels within 2 km of the outfall (Delille and Delille Reference Delille and Delille2000). Not surprisingly, given the large numbers (over 1000) of people working at the U.S. McMurdo Station, this station's outfall has a large plume, 200–300 m wide and 1 km long, with coliform bacteria densities up to 105/100 ml (McFeters and others Reference McFeters, Barry and Howington1993). As yet these plumes have not been shown to cause other than localised changes but there remain concerns about the possible transmission of diseases to wildlife. This issue was dealt with in a paper prepared by a SCAR and COMNAP Working Party (SCAR and CONMAP 2000). It states that no disease outbreak in Antarctica positively linked to human activities has yet been identified but recommends the taking of precautions to obviate such occurrences.
A third driver for change has been the realisation by a number of countries that sewage discharges of the magnitude and nature of those at their research stations would require more extensive treatment if they occurred at home given regulatory requirements in their respective countries. In other words they elected to regard their Antarctic bases as coming under the same requirements as would equivalent sized populations within their countries. This was one of the factors contributing to the decision to install a new treatment plant at New Zealand's Scott base (New Zealand Reference Zealand2002).
Design considerations
Wastewater flowrates and composition
There is a remarkable consistency between the per capita design flowrates used in sizing recently constructed or planned wastewater treatment plants at medium sized research stations. For the similar plants at the German Neumayer and the Argentine Jubany stations, flowrates of 145 and 133 l/day/capita were used (Ahammer and others Reference Ahammer, Matz, El Naggar, Gernandt and Shirashi2000). Very similar values of 164 and 160 l/day/capita respectively were employed for New Zealand's new Scott base plant (New Zealand Reference Zealand2002) and that planned for Australia's Davis station (Heaton and Paterson Reference Heaton and Paterson2003), while for the Japanese Syowa station plant, the corresponding figure was 150 l/day/capita (Umezawa and others Reference Umezawa, Sakamoto, Yanagiya, Yoshida, Kato and Shiraishi2000). These authors point out that the latter figure is only 60% of the flowrate used in Japan itself; this reflects the high energy costs associated with water production at Antarctic research stations and the resulting emphasis on water conservation.
A key design consideration at all plants is the marked seasonal variation in hydraulic loading rates. The number of people overwintering on bases can be as few as 10% of the peak number present during the summer resupply period. Actual or projected minimum and maximum numbers for various bases are: Davis (Australia): 20 and 150 (Heaton and Paterson Reference Heaton and Paterson2003); Scott (New Zealand): 12 and 120 (New Zealand Reference Zealand2002); and Rothera (U.K.): 20–25 and 100+ (Hughes and Blenkharn Reference Hughes and Blenkharn2003).
It is recognised that per capita BOD loads in Antarctica tend to be high. Lori and others (1996) cite a figure of 100–120 g BOD per person, contrasting this with the figure of 60 g BOD per person often quoted in textbooks. This high per capita BOD level can be attributed to the fact that people working in Antarctica are well fed and have a high calorific value food input rich in fats; there is also a predominance of males, who produce more wastes than females (Heaton and Paterson Reference Heaton and Paterson2003). It is also possible that the higher per capita BOD at research stations might be caused by a more concentrated waste stream. In addition to human waste, municipal wastewater in a typical city also contains grey water from washing activities and in some cases wastewater may also be diluted with ground water infiltrating into sewer lines. The addition of grey water and in some cases infiltrating ground water to the waste coming from human contribution would result in a lower per capita BOD measured in the influent at a wastewater treatment facility. It would be expected that the wastewater stream in most research stations in Antarctica is concentrated with little dilution from washwater since water use is restricted.
Although per capita BOD outputs would not be expected to differ much between bases, design BOD levels show substantial variations, and this is true also of suspended solids (SS) levels; for example in designing the Scott base plant SS and BOD levels of 700 mg/l were used (New Zealand Reference Zealand2002) while in the design of the Japanese Syowa station SS and BOD levels of 320 mg/l were assumed (Umezawa and others Reference Umezawa, Sakamoto, Yanagiya, Yoshida, Kato and Shiraishi2000).
An additional complication at coastal stations using seawater for toilet flushing is the high salt content of the wastewater: at Scott base up to 28% of the sewage is seawater (New Zealand Reference Zealand2002).
Effluent specifications and sludge disposal criteria
Treatment plant designs are also crucially dependent upon effluent specifications and on how any sludges produced are to be dealt with. Here there are significant differences between countries. For example, at the plants of German design installed at the Neumayer (1997) and Jubany (1998) bases an effluent BOD of 50 mg/l, an SS of 100 mg/l and a faecal coliform target of 200 cfu/100 ml were specified (Ahammer and others Reference Ahammer, Matz, El Naggar, Gernandt and Shirashi2000), while for their Scott base plant (2002) the New Zealanders adopted the same coliform target but a more demanding 30 mg/l standard for both BOD and SS (New Zealand Reference Zealand2002). An even more demanding 20 mg/l BOD, 30 mg/l SS target was set by the Australians for their planned plant at Davis, with complete disinfection of the effluent being aimed at (Heaton and Paterson Reference Heaton and Paterson2003). Over time, effluent specifications seem to be becoming more exacting, a trend consistent with what is happening elsewhere in the world.
As far as sludges are concerned there seems to be unanimity across all countries that have built or are planning new treatment plants that sludges and screenings should be removed from the Antarctic Treaty area. This generally means that plants need to incorporate sludge dewatering facilities so as to minimise the effort and expense involved in transporting these elsewhere.
Other design considerations
Extreme conditions
Climatic conditions in Antarctica are severe. The yearly temperature range of –40°C to+5°C experienced at Davis, with wind gusts of up to 200 km/h (Heaton and Paterson Reference Heaton and Paterson2003), would be not untypical of conditions at coastal bases on the Antarctic continent. Inland, temperatures are much lower. At the Amundsen-Scott South Pole station, one of only two permanently occupied inland bases, temperatures range from –83°C to –14°C, with a mean of –50°C, while the wind blows constantly, with an average speed of 20 km/h (Flynn and others Reference Flynn, Bubenheim, Straight and Belisle1994). Blizzards are not infrequent and can develop very quickly; snowfalls may not be large (78 mm annually at Davis (Heaton and Paterson Reference Heaton and Paterson2003)) but drifts of windblown snow accumulate around prominent objects such as buildings. Freezing of pipes connecting plant components can be a serious problem (Lori and others Reference Lori, Meneghello, Scarano, Voli, Melander and Fontana1992). It also means that treated effluent discharge is usually best done intermittently so as to minimise the chance of outfall pipes freezing up. This in turn implies that plants need to include a buffer tank in which treated effluent can be stored until the next discharge.
Construction and transport
The climate, together with the prolonged period of darkness during winter and the fact that stations maintain only a skeleton staff during the winter months, means that there exists a total of perhaps eight weeks in summer when conditions are good for outdoor construction work (Heaton and Paterson Reference Heaton and Paterson2003). Much of this time will be needed for transport of plant components to the selected site, installation, and connection of pipes and utilities. Modular, prefabricated plants that can be easily disassembled for shipping and readily reassembled on site are therefore virtually a necessity. Such plants can also be commissioned and modified prior to shipping, further alleviating demands on the limited staff available on-site at research stations. For convenience of transport and handling, and to protect plant components during the rough weather characteristic of the southern ocean, most countries are opting to build their plants inside containers. The Scott base plant fits inside two ISO-20 shipping containers (New Zealand Reference Zealand2002) while the Neumayer and Jubany plants were built into containers 2.0 m long, 1.8 m wide and 1.8m high, made of stainless steel to minimise problems with corrosion (Ahammer and others Reference Ahammer, Matz, El Naggar, Gernandt and Shirashi2000).
No matter what form they take, treatment plants still need to be installed on appropriate support frames and foundations. Given the conditions in Antarctica, these are not always simple to design and construct. For example, if concrete is to be used, aggregate is needed. Shipping in aggregate is expensive and inconvenient; however, on-site quarrying and crushing is noisy, produces dust and creates a permanent scar on the landscape. It can also disturb wildlife. At Casey station, for example, the quarry is close to a snow petrel breeding colony while Wilson's petrels breed beneath exposed rocky outcrops all around the station. If the spirit of the Madrid Protocol is to be observed, this effectively precludes use of the quarry during the summer months, the only time when it is readily accessible.
Design and operation
Compactness is necessary if plants are to fit into containers. It also has other advantages, reducing the plant's footprint and its visual impact (Heaton and Paterson Reference Heaton and Paterson2003) as well as minimising heating costs. Rates of microbial waste decomposition processes are temperature dependent and to date plants have all been designed to run at 12°C (New Zealand Reference Zealand2002) or above.
Plants must also be able to handle widely varying hydraulic and BOD loads since, as discussed earlier, the number of people overwintering at research stations is a small fraction of those present during the peak summer resupply period. Most people present on bases have similar work schedules so diurnal load fluctuations are pronounced and have to be allowed for. Another aspect of importance is that numbers on stations tend to alter stepwise, for example when the resupply vessel arrives. This means that plants have to be able to adjust rapidly to the load changes. This problem is most acute at seasonally occupied stations. At Spain's Juan Carlos I base, used only between December and February each year, the sewage treatment facility took two weeks to recommission, causing serious problems in the interim (Alva Reference Alva and Shirashi2000).
Winter weather conditions, together with the small number of people present at some times of year, means that plants need to be robust, highly dependable, simple to operate and requiring little maintenance. The remoteness and isolation of plants means that spare parts may be unavailable for long periods and therefore plants must incorporate a high level of redundancy for all items of equipment not readily repairable on site; this approach was followed, for example, at the Neumayer and Jubany stations and at Scott base (Ahammer and others Reference Ahammer, Matz, El Naggar, Gernandt and Shirashi2000; New Zealand Reference Zealand2002). Particularly during winter, plant operation is usually handled by trades staff such as the station plumber (Heaton and Paterson Reference Heaton and Paterson2003) who usually has numerous other responsibilities (Stewart and others Reference Stewart, Brookman and Rodionov2003) and lacks a detailed understanding of wastewater treatment fundamentals. This implies that a highly automated control system is desirable, yet it must still be possible to run the plant on manual control in the event of a control system breakdown (Stewart and others Reference Stewart, Brookman and Rodionov2003).
Plants need also to have low energy consumption. Energy is expensive in Antarctica, being derived from hydrocarbon fuels shipped in at high cost and stored in heated storage tanks. Aside from saving money, installing a low power demand plant means a reduced output of emissions from the base's power station, a desirable outcome anywhere but especially in Antarctica.
As well as reducing emissions, it is also important to prevent odours occurring either within station buildings or around the station precincts. This requires avoidance of anaerobic conditions, and the enclosing and ventilating of areas where raw sewage and primary solids are present. Where the plant is partly heated by air drawn from other buildings, backflow of air needs to be prevented, for example by running the plant at a slight negative pressure (Stewart and others Reference Stewart, Brookman and Rodionov2003).
Evolution of treatment plant designs
Many countries active in Antarctica have still made no move to take sewage treatment beyond the minimum standards prescribed by the Madrid Protocol (Hughes Reference Hughes2004). However, others have elected to install more advanced treatment systems, usually incorporating biological treatment and sludge dewatering processes. According to Heaton and Paterson (Reference Heaton and Paterson2003), Australia was the first country to decide to introduce secondary treatment at its Antarctic stations. This decision was taken well before the Madrid Protocol was drafted, with Rotating Biological Contactors (RBCs) being installed at Australia's Mawson, Casey and Davis stations in 1985, 1989 and 1991 respectively. During this period several other countries also introduced biological treatment plants, all based on RBCs. The Italians installed their plant at their Terra Nova base in 1987–1988. This base is not permanently occupied and problems were experienced due to the time taken to get the RBC system functioning properly at the start of each summer season (Lori and others Reference Lori, Meneghello, Scarano, Voli, Melander and Fontana1992). India also introduced two RBC systems at this time, one in 1988 and one in 1989. These were installed at their permanent research station of Maitri (Ghosh and others Reference Ghosh, Muley, Ghode and Ramteke1997). The discs used on these plants were rather unusual, being mostly square in shape.
The Australian plants comprise a flow equalisation tank, a primary sedimentation tank, an RBC with six banks of discs in series (to help cope with the widely varying seasonal loads), a clarifier and an effluent holding tank (Heaton and Paterson Reference Heaton and Paterson2003). All three are still in use and those at Mawson and Casey appear still to be working satisfactorily, though no detailed analysis of their performance appears to be available. Some problems have been reported at Davis where the RBC was not constructed in its own building as was the case at Casey and Mawson. According to Heaton and Paterson (Reference Heaton and Paterson2003), this resulted in access difficulties over winter and a consequent inability to carry out maintenance activities. Poor ventilation, resulting in the system becoming anaerobic was also noted as a problem. These problems provided the stimulus for the design of a new plant at Davis, as described in Heaton and Paterson (Reference Heaton and Paterson2003).
Experience at the Italian and Indian plants mentioned above was also not very encouraging. Lori and others (Reference Lori, Meneghello, Scarano, Voli, Melander and Fontana1992) noted that the BOD removal efficiency achieved by the Terra Nova plant was low, but this could well be due to the plant's lengthy start-up time. As Lori and others (Reference Lori, Meneghello, Scarano, Voli, Melander and Fontana1992) recognised, the time needed to re-start biological treatment plants casts serious doubt on their suitability for use at seasonally occupied stations. Low BOD removal efficiencies, of 25 to 40%, were also reported for the Maitri plant and these were associated with significant falls in effluent pH (Ghosh and others Reference Ghosh, Muley, Ghode and Ramteke1997). Reasons for this poor performance were not given but observations that at times the discs were not rotating suggests that the plant was experiencing mechanical problems, resulting perhaps in the development of anaerobic conditions in the RBC tank.
Whilst biofilm-based plants have predominated at Antarctic research stations, a suspended growth design was introduced in the late 1990s at the German Neumayer station and the Argentine Jubany base. The design was developed from an already well-established technology approved by the German maritime employees’ liability insurance association. In the plant a continuous flow of wastewater is drawn from upstream equalisation and buffering tanks and fed to the first of two aerated chambers in series. Suspension from the second chamber passes to a third chamber which operates cyclically; it is aerated when it is desired to recycle sludge back to the first chamber but for the remainder of the time it serves as a clarifier. At the end of each settling phase the supernatant is pumped to a fourth chamber and thence through a UV reactor to the outfall. Excess sludge is drawn from all of the first three chambers and processed as described later. The plants are heavily automated, require only around 6–8 hours of supervision per month and, as of 2000, had been operating continuously for 3 years (Ahammer and others Reference Ahammer, Matz, El Naggar, Gernandt and Shirashi2000).
Despite the success of the German design, all recent plants appear to have stayed with biofilm-based systems. No documented discussion of the reasons why RBCs were chosen for the early plants appears to be available. However, detailed analyses of possible technologies and their respective advantages and disadvantages have been provided for the planned new plant at Davis (Heaton and Paterson Reference Heaton and Paterson2003), the recently commissioned plant at Scott base (New Zealand Reference Zealand2002), and, to a lesser extent, for the Japanese Syowa station plant (Umezawa and others Reference Umezawa, Sakamoto, Yanagiya, Yoshida, Kato and Shiraishi2000).
What is interesting is that in all three cases RBCs and contact aeration systems employing submerged aerated fixed media were selected as the two most suitable technologies. New Zealand opted to go for contact aeration as they had concerns about drive shaft problems and the height requirements of the RBCs. The latter would make it hard to construct the plant within a standard container while still retaining easy access to all parts of it (New Zealand Reference Zealand2002). The Japanese made a similar choice, as have the Americans at their McMurdo station and the British at Rothera (Umezawa and others Reference Umezawa, Sakamoto, Yanagiya, Yoshida, Kato and Shiraishi2000; Hughes Reference Hughes2004). The Australians, however, elected to stay with RBCs for Davis, although they plan to introduce some significant modifications: discs are to be replaced by cylindrical drums containing spiral media while the contactors are to be made from buoyant material so as to overcome some of the mechanical problems that have afflicted past RBCs. The plant dimensions will be 8.2 m long, 2.27 m wide and 2.9 m high (Heaton and Paterson Reference Heaton and Paterson2003), precluding its construction within a standard container.
Whilst almost all treatment plants proposed for use in Antarctica have used biological processes, the unsuitability of these processes for use at seasonally occupied bases prompted the Italians to try out a physico-chemical plant at their Terra Nova base (Lori and others Reference Lori, Meneghello, Scarano, Voli, Melander and Fontana1992). This plant comprised a wastewater storage tank, a chemical reactant preparation and storage section, a flocculation reactor, a dissolved air flotation (DAF) unit, a final effluent receiver and a sludge processing unit. The process required sequential injection of a pH corrector, a primary coagulant and a polyelectrolyte (Lori and others Reference Lori, Meneghello, Scarano, Voli, Melander and Fontana1992). Difficulties were experienced when trying to commission this process and later the decision was taken to combine it with the RBC system it was designed to replace. The resulting hybrid plant comprises newly added screening equipment, the flocculation and DAF units from the physico-chemical plant, followed by the RBC unit from the earliest plant. To these were added an activated carbon filter and a UV disinfection unit. The sludge drying unit from the physico-chemical plant was also retained. This hybrid plant was reported to be performing well; inclusion of the physico-chemical section was noted as having helped improve treatment efficiencies when the plant was started up (Lori and others Reference Lori, Ponzo, Indulti, Stefanoni and Hall1996).
No matter what type of process has been selected, all recent plants have opted to disinfect their effluent prior to discharge. In all cases UV radiation has been the process of choice since it leaves no potentially harmful residuals. As Heaton and Paterson (Reference Heaton and Paterson2003) point out, however, UV lamps of medium-strength or stronger are required for the cooler temperatures found in Antarctic treatment plants. Because the presence of suspended particles in the effluent can lead to inefficient disinfection, they recommend the use of multiple pass systems preceded by a filter.
Whilst there is much consistency as far as the choice of secondary processing technologies is concerned, a wide diversity of sludge processing routes has been employed. All countries that have taken the trouble to construct modern treatment plants seem to have accepted that sludge should be repatriated for final disposal. To minimise costs most sludges are dewatered but the final solids content achieved and the methods employed vary greatly.
At Italy's Terra Nova base a filter press was used (Lori and others Reference Lori, Ponzo, Indulti, Stefanoni and Hall1996) while at America's McMurdo Base a belt filter is employed; both these yield a sludge containing around 25% solids. Much more extreme approaches are used at the German and Japanese plants. At the former, sludge is treated with lime and flocculant, dried in an oven, then stored in bags before being shipped out; dry sludge production rates are around 4 kg per person per year (Ahammer and others Reference Ahammer, Matz, El Naggar, Gernandt and Shirashi2000). At Japan's Syowa plant, polymer and coagulant are added, the sludge is put through a pressure filter and then reportedly ‘carbonised’ before being returned to Japan (Umezawa and others Reference Umezawa, Sakamoto, Yanagiya, Yoshida, Kato and Shiraishi2000). Yet another approach is used at New Zealand's Scott Base where sludge is centrifuged and the dewatered product stored in drums before being shipped home. Only at Davis does there seem to be no dewatering, with the plan being to ship out as-produced sludge (4% solids) (Heaton and Paterson Reference Heaton and Paterson2003). This will probably be only a temporary measure since prospects for using the freeze-thaw method for sludge dewatering are being actively followed up by the Australian Antarctic Division. This is a promising approach that has undergone pilot scale testing in the USA, while a detailed plan for a freeze-thaw sludge dewatering facility at McMurdo station has been developed (Martel Reference Martel2001). In addition preliminary tests at Casey station have yielded a 46% solids product from a 2.5% solids sludge feedstock (Heaton and Paterson Reference Heaton and Paterson2003).
Concluding remarks
It is evident that the level of sewage treatment applied at Antarctic research stations varies considerably. Many countries have decided to do no more than is required of them by the Madrid Protocol whereas others have taken the decision to treat their sewage as they would back home. It is, as yet, unclear what long term benefits the use of more advanced treatment facilities will bring to Antarctica. Certainly the levels of faecal coliforms in the vicinity of outfalls can be markedly reduced, as was demonstrated at the British Rothera station (Hughes Reference Hughes2004). However, what this means for local ecosystems remains uncertain.
Despite these uncertainties, it is expected that increasing numbers of countries will embrace the precautionary principle and decide to install treatment plants at their Antarctic bases. Most will adopt technologies similar to those currently in use but, at inland stations particularly, innovative uses of high-tech membrane and thermal evaporative techniques are to be expected. Not all innovation needs to involve high-tech processes however; for example, Hughes and Blenkharn (Reference Hughes and Blenkharn2003) have shown that replacement of intermittent effluent tank discharging by continuous cold water flushing can bring about a 90% reduction in faecal coliform concentrations around outfalls.
Whatever path sewage treatment in Antarctica takes, it is certain that there will be no lack of interesting challenges for those with responsibility for design, construction and operation.
Acknowledgements
The help of the Australian Antarctic Division (AAD) in making it possible for me to spend the 2000–2001 summer at Casey station is much appreciated. The help received from various AAD staff members in collecting material used in this paper is also gratefully acknowledged.
